Data Prep

This workspace contains a batch converter that reads files from ./data/ and writes Markdown files to ./output/. The converter preserves subdirectories, so data/กฎหมายดิจิทัล/files becomes output/กฎหมายดิจิทัล/files.

Supported extensions:

• pdf
• pptx
• docx
• xlsx
• xls

Setup

Use Python 3.12 or later (other versions may work fine)

Install the Python dependencies:

pip install -r requirements.txt

Create a .env file in the project root based on .env.example and add your Typhoon API key (for PDF file conversion).

Usage

Put the files you want to convert inside ./data/, then run:

python typhoon_ocr.py

If a mirrored output path already exists, the script uses numbered suffixes starting from _1, _2, and so on.

The generated Markdown files will appear in ./output/ as filename.md